Pemodelan Banyaknya Kasus Kematian COVID-19 di Jawa Timur Menggunakan Negative Binomial Regression dan Generalized Poisson Regression

Fikansa, Fimadasa Blesofi (2023) Pemodelan Banyaknya Kasus Kematian COVID-19 di Jawa Timur Menggunakan Negative Binomial Regression dan Generalized Poisson Regression. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06211940000142-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (5MB) | Request a copy

Abstract

Coronavirus disease-2019 (COVID-19) adalah penyakit yang disebabkan oleh virus severe acute respiratory syndrome coronavirus 2 (SARS-COV2). Perkembangan virus yang cukup pesat menyebabkan virus tersebut cepat menyebar di Indonesia, bahkan Indonesia pernah menjadi negara dengan tingkat kematian tertinggi akibat COVID-19 di Asia Tenggara pada tahun 2021. Provinsi Jawa Timur menjadi provinsi dengan kasus kematian terbanyak akibat COVID-19 pada tahun 2021 dengan persentase kematian sebesar 7,57%. Ada beberapa faktor yang diduga mempengaruhi kasus kematian COVID-19 di Jawa Timur, diantaranya adalah persentase penduduk lansia, persentase kasus penyakit tuberculosis, persentase penyakit diabetes melitus, persentase tenaga dokter, persentase tenaga perawat, dan persentase fasilitas kesehatan. Pemodelan banyaknya kasus kematian COVID-19 di Jawa Timur dilakukan menggunakan pendekatan metode Generalized Poisson Regression (GPR) dan Negative Binomial Regression (NBR). GPR dan NBR adalah metode yang digunakan ketika terjadi overdispersi pada Regresi Poisson. Pemodelan juga dilakukan dengan menggunakan exposure, dimana exposure adalah variabel yang diprediksi dapat mempengaruhi sebuah kejadian. Pada penelitian ini, exposure yang digunakan adalah jumlah kasus positif COVID-19 di setiap Kabupaten/Kota di Jawa Timur. Berdasarkan analisis yang telah dilakukan, diperoleh hasil bahwa jumlah kasus kematian terbanyak terjadi di Kabupaten Blitar sebanyak 1.489 kasus dan terendah adalah Kabupaten Sampang sebanyak 103 kasus. Kriteria AICC menunjukkan bahwa metode Regresi Generalized Poisson dengan exposure merupakan model terbaik karena memiliki nilai AICC yang lebih kecil dibandingkan metode Regresi Poisson maupun Regresi Binomial Negatif. Variabel yang mempengaruhi jumlah kematian akibat COVID-19 di Jawa Timur secara signifikan adalah persentase penduduk lansia dan rasio penyakit diabetes melitus.
===============================================================================================================================
Coronavirus disease-2019 (COVID-19) is a disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-COV2) virus. The rapid development of the virus caused the virus to spread quickly in Indonesia. In fact, Indonesia was once the country with the highest death rate due to COVID-19 in Southeast Asia in 2021. East Java Province became the province with the most cases of deaths due to COVID-19 in 2021 with the percentage of deaths being 7.57%. There are several factors that are suspected of influencing the COVID-19 death cases in East Java, including the percentage of the elderly population, the percentage of tuberculosis cases, the percentage of diabetes mellitus, the percentage of doctors, the percentage of nurses, and the percentage of health facilities. Modeling the number of COVID-19 death cases in East Java was carried out using the Generalized Poisson Regression (GPR) and Negative Binomial Regression (NBR) methods. GPR and NBR are methods used when overdispersion occurs in Poisson Regression. Modeling is also done using exposure, where exposure is a variable that is predicted to affect an event. In this study, the exposure used was the number of positive cases of COVID-19 in each Regency/City in East Java. Based on the analysis that has been carried out, the result is that the highest number of death cases occurred in Blitar Regency with 1,489 cases and the lowest was in Sampang Regency with 103 cases. The AICC criteria show that the Generalized Poisson Regression method with exposure is the best model because it has a smaller AICC value than the Poisson Regression or Negative Binomial Regression methods. The variables that significantly affect the number of deaths from COVID-19 in East Java are the percentage of the elderly population and the ratio of diabetes mellitus.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	COVID-19, Exposure, Overdispersi, GPR, NBR, Overdispersion
Subjects:	R Medicine > RA Public aspects of medicine > RA644.C67 COVID-19 (Disease)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Fimadasa Blesofi Fikansa
Date Deposited:	05 Sep 2023 08:28
Last Modified:	05 Sep 2023 08:28
URI:	http://repository.its.ac.id/id/eprint/104470

Actions (login required)

View Item